Risk Assessment Outline

1 Risk Assessment Outline I. Why do risk assessments? A. Help owners make decisions about aggressive animals B. Help shelters/rescues decide what animals to adopt to the public C. Help courts decide legal issues surrounding animals II. III. IV. The scientific and rational basis for risk assessment A. Once an animal has injured, the probability of future injuries is increased B. Studies examining risk factors Factors to consider in assessments A. Animal factors B. Potential victim factors C. Environmental factors D. Owner/family factors How to do an assessment A. Behavioral history B. Incidental direct observations C. Formal testing V. Client communication Talking to owners VI. VII. Risk assessment in the shelter/rescue setting Risk assessment in legal settings

2 Risk Assessment Lecture Notes Aggressive behavior is one of the most common behavior problems seen by behavior consultants (Borchelt & Voith, 1985). It is also one of the most serious of behavior problems involving dogs. It is estimated that there are more than 60 million dogs in the U.S at present and each year, about 4.8 million people are bitten by dogs. Of these, over 300,000 seek medical help for their bites and between 10 and 20 people are killed as a direct result of dog bites (Reviewed in AVMA Task Force, 2001). For up to date information on dog bites and prevention, go to the Centers for Disease Control and Prevention National Center for Injury Prevention and Control website at http://www.cdc.gov/ncipc/duip/biteprevention.htm. Aggression tends to have different definitions to different people, creating confusion, and dogs are often said to be aggressive without identification of the criteria for labeling the dog in that way. Identifying a dog as vicious or dangerous suffers from the same definitional problems. Aggression can be defined as behavior that causes harm to another with intent to cause harm. The term vicious has no meaning in animal behavior but sometimes has legal meanings in ordinances and laws. On one level it is not dogs that are aggressive, vicious or dangerous it is the behavior that has these characteristics. When a dog is classified as aggressive, vicious or dangerous it is often implied that the dog has shown aggressive behavior that puts people or animals at risk of injury or death. How much of this behavior must be shown to classify a dog as aggressive or dangerous is often not clearly specified. Risk assessment is an evaluation of an animal s likelihood of causing injury or death to others. It may be based on historical information about the animal and his environment and/or on behavioral observations. Risk assessments may be done as part of a behavioral consultation for a client with an animal that has a behavioral problem, as part of an evaluation of an animal in a shelter or rescue setting or as part of an evaluation of an animal for legal proceedings. Risk assessment may be part of a prognosis for improvement of the behavior but it is not the same as a prognosis. A prognosis is a prediction of the probable course and outcome of a medical disease or behavioral problem (Blood & Studdert, 1999). A prognosis may look at other things such as age of onset of the problem, whether it is increasing or decreasing in frequency or intensity, the complexity of the problem and the occurrence of other problems. See Landsberg, et al., 2003, pp 85-87 for a discussion of prognosis of behavior problems. Often a poor risk assessment indicates a poor prognosis for the problem. The scientific and rational basis for risk assessment At present, there is not a strong empirical basis for most risk assessment. There are very few studies that examine or identify risk factors for aggression in animals. Most risk factors come from rational arguments. Most studies suggesting risk factors are weak suffering from problems with sampling subjects or aggressive incidents. As a result it is difficult to draw strong conclusions about things that increase the risk of injury or death by an animal. For example, many studies that argue that members of certain breeds of

3 dogs are more dangerous than others (there is a higher risk of injury from certain breeds than others) have serious problems with sampling and breed identification. Some risk factors have been suggested based on rational rather than empirical arguments such as: big dogs have big jaws and are therefore capable of doing more damage than small dogs with small jaws. Therefore, large dogs are more dangerous than small dogs. Risk factors for dog aggression suggested by the literature Being the member of certain breeds: Such as Pit bull, Rottweiler, German Shepherd Being a male dog Being an intact male dog Being large > 50 lbs Having certain medical conditions: Rabies, hip dysplasia Having children present in the household Having exposure to very young or very old persons or those that are disabled Having exposure to human males <10 years of age Poor socialization Factors to consider in any risk assessment. A variety of people have suggested factors to be evaluated in doing a risk assessment. The ones listed here are a composite of several sources including Hetts and, Landsberg, et al. We have grouped the factors into four categories: animal related factors, victim related factors, environmental factors and owner/family factors. All should be considered in any risk assessment. A. Animal factors 1. Kind of injury delivered 2. Identifiably of triggers and situations that lead to aggression 3. Consistency of response to triggers 4. Benign-ness of triggers or over-reactivity 5. Latency to respond aggressively 6. Existence and reliability of warning signals 7. Number of injuries caused over time 8. Severity of injuries caused 9. Size and strength of the animal 10. Ease of Interruption 11. Latency to return to non-aggressive motivational state B. Potential victim factors 1. Vulnerability of victims (young, old, disabled) 2. Consistency of target (only one person or anybody) 3. Ability of victims to identify threats and aggression 4. Ability and motivation to avoid aggressive situations C. Environmental factors 1. Number and variety of situations and triggers that elicit aggression

4 2. Safety of the environment D. Owner/family factors 1. Comprehension of potential danger 2. Ability and motivation to provide safe management 3. Control over animal 4. Family size, consistency, lifestyle 5. History of responsible pet ownership 6. Overall history of responsible citizenship How to do a risk assessment. Most risk assessments consist of gathering historical information from the owner or keeper of the animal about the animal s past and present behavior as well as direct observations of the animal. Historical information may also include information from others that have interacted with the animal (neighbors, victims, veterinarians, trainers, shelter kennel staff), medical records, animal control records and legal documents. Taking a good history is absolutely essential. History-taking and interview techniques are described in references such as Hetts (1999) and Landsberg, et al. (2003). In some settings such as shelter or rescue operations, a history may be fragmentary or non-existent. In those situations, direct observations become more important. Observations of the animal can be made incidentally to an interview of the owner/keeper or could be more formal involving a specific testing protocol. Direct observations should be made whenever possible. A variety of formal testing protocols are available including those of Sternberg, Weiss, and Hetts & Estep which were designed to more broadly assess the suitability of dogs for adoption. Problems with validity and reliability are present in most formal testing protocols. Client communication Owning a potentially dangerous animal is distressing to most people. Considering euthanizing a dangerous animal is also distressing to owners. Talking to distressed people is difficult and must be done carefully if the owners are to hear what you have to say. Owners do not need to be told what to do. They need impartial information to make their own decisions about what to do with the animal. The same is true of shelters and rescue organizations that assess the risks of adopting out a particular animal. The decisions are often complex, and what would be perceived as a high risk animal by one owner, shelter or rescue may be considered a lower risk animal by others. Risk assessment in the shelter/rescue setting Risk assessment of shelter or rescue animals is often part of a larger evaluation of the animal for potential adoption. An animal may appear to be a low risk for injury to others but not be considered adoptable for other medical or behavioral reasons. An example might be a cat that is very friendly but has a urine marking problem. Many of the animal factors listed above should be considered in evaluations of animals for adoption. Victim, environmental and owner/family factors should be considered in pairing the animal with potential adopters.

5 Risk assessment in legal settings Sometimes attorneys, courts or other legal entities may request an evaluation of an animal as to his potential dangerousness to others. Procedures for evaluation are similar to those for evaluating an individually owned animal and the same factors should be considered. Depending upon the legal circumstances, there may be limitations on what can be evaluated and how it can be evaluated. For example, a dog on quarantine may not be allowed out of his enclosure at an animal control facility, making some evaluation procedures difficult or impossible. Thought Questions 1. How is a risk assessment different from a temperament evaluation or a prognosis of a behavior problem? 2. Of the environmental factors to be considered in a risk assessment, what does safety of the environment refer to? 3. Is a risk assessment the same thing as an evaluation of a dog for adoption? If not, how are they different? 4. What kind of informal observations can be made of dogs at a shelter that might contribute to a risk assessment?

6 Risk Assessment References and Suggested Readings AVMA Task Force on Canine Aggression. 2001. A community approach to dog bite prevention. Journal of the American Veterinary Medical Association, 218, 1732-1749. Available at http://www.avma.org/public_health/dogbite/dogbite.pdf Blood, D.C. & Studdert, V.P. 1999. Saunders Comprehensive Veterinary Dictionary, 2 nd Ed. New York, NY: W.B. Saunders. Borchelt, P.L. & Voith, V.L. 1985. Aggressive behavior in dogs and cats. Compendium for Continuing Veterinary Education, 3, 730-738. Gershman, K.A., Sachs, J.J. and Wright, J.C. 1994. Which dogs bite? A case-control study of risk factors. Pediatrics, 93, 913-917. Hetts, S. 1999. Pet Behavior Protocols. What to Say, What to Do, When to Refer. Lakewood, CO: AAHA Press, pp.169-202. Landsberg, G., Hunthausen, W. and Ackerman, L. 2003. Handbook of Behavior Problems of the Dog and Cat, 2 nd Ed. New York: Saunders, pp.388-390. Lockwood, R. 1994. The ethology and epidemiology of canine aggression. In Serpell, J. ed. The Domestic Dog: It s Evolution, Behavior and Interactions with People. New York, NY: Cambridge University Press, pp.131-138. Wright, J.C. 1996. Canine aggression: Dog bites to people. In Voith, V.L. & Borchelt, P.L. eds. Readings in Companion Animal Behavior. Trenton, NJ: Veterinary Learning Systems, pp.240-246. Adoption Assessment Protocols Hetts, S. and Estep, D.Q. 2000. Canine Behavior, Evaluating Behavioral Health. (DVD program) Denton, TX: Animal Care Technologies. Available at AnimalBehaviorAssociates.com Sternberg, S. Assess-A-Pet : The Manual. Available at SueSternberg.com Weiss, E. The Meet-Your-Match Safer Assessment. Available at ASPCA.org.

7 Some Studies of Risk Factors for Dog Aggression Daniel Q. Estep, Ph.D., CAAB Animal Behavior Associates, Inc. These studies illustrate the diversity of methods, definitions and results that have been found. This is not a complete list of studies looking at dog bite injuries but a selected sample. Examining them cautions us about drawing general conclusions about the causes of dog bites. Gershman, K.A., Sacks, J.J. & Wright, J.C. 1994. Which dogs bite? A case-control study of risk factors. Pediatrics, 93, 913-917. There were 991 dog bites reported to animal control in Denver, CO in 1991. Of these they select ones that were a first bite by the dog to a non-family member that received medical treatment, and that the dog was the only one involved in the incident and had been owned more than 6 months. These 178 cases were compared to a matched sample of dogs living in the same area of town that had not bitten. Compared to controls, biting dogs were more likely to be German shepherd dogs or Chow-Chows, were more likely to be intact males, more likely to reside in a home with more than one child and more likely to be chained in the yard. The majority of victims were males, less than 12 years old. The majority of bites were considered minor rather than severe. Guy, N.C., Leuscher, U.A., Dohoo, S.E., Spangler, E., Miller, J.B., Dohoo, I.R. & Bate, L.A. 2001. Risk factors for dog bites to owners in a general veterinary caseload. Applied Animal Behavior Science, 74, 29-42. Surveyed 515 veterinary clients in Canada, 227 said they currently owned a dog that had bitten at least once, even if an accident. Mixed breed dogs were the most common biters although golden retrievers were the most common pure bred biters. Risk factors were being a female dog, being small, having a history of skin diseases and having teenagers in the home. Victims were just as likely to be males as females but more likely to be adults than children. Less than 10% sought medical attention. Sacks, J.J., Sattin, R.W. & Bonzo, S.E. 1989. Dog-bite related fatalities from 1979 through 1988. Journal of the American Medical Association, 262, 1489-1492. Gets data for dog-bite related fatalities from newspapers as well those from medically reported deaths. Risk factors for death include: being attacked by a pit-bull, being a child less than 10 years old. Sacks, J.J., Sinclair, L., Gilchrist, J., Golab, G.C. & Lockwood, R. 2000. Breeds of dogs involved in fatal human attacks in the United States, 1979-1998. Journal of the American Veterinary Medical Association, 217, 836-840. Gets data for dog-bite related fatalities from newspapers as well those from medically reported deaths. Shows one risk factor, breed of dog, changes over time, and there are serious problems with identifying a particular breed as more likely to kill than others.

8 Simpson, B.S. 1997. Aggression Dogs. In Tilley, L.P. & Smith, F.W.K. eds. The Five-Minute Veterinary Consult: Canine and Feline. Baltimore, MD: Williams & Wilkins, pp.10-11. Provides a list of questions (risk factors) that should be examined with any client presenting an aggressive dog. Factors come from other studies, such as those above, and rational arguments. Wright, J.C. 1985. Severe attacks by dogs: Characteristics of the dogs, the victims and attack settings. Public Health Reports, 100, 55-61. Examined 5,711 bites in five counties over 3 years. Identified 16 as severe. Severe attacks were where the dog repeatedly bit or vigorously shook the victim and there was extreme difficulty terminating the attack. Most common breed was American Staffordshire Terrier, all were intact males, 8 were chained or broke chains, most had a previous history of aggression, average age of victim was 8 years, most victims were family members or acquaintances. For critiques of the methodology and conclusions drawn from these studies, see: Bradley, J. 2005. Dogs Bite, But Balloons and Slippers Are More Dangerous. Berkeley, CA: James and Kenneth Publ. Delise, K. 2007. The Pit Bull Placebo. The Media, Myths and Politics of Canine Aggression. Anubis Publishing.

9 CANINE AGGRESSION I: THE VETERINARIAN S ROLE IN ADVISING CLIENTS Suzanne Hetts, Ph.D. Certified Applied Animal Behaviorist Animal Behavior Associates, Inc. 4994 S. Independence Way Littleton, CO USA www.animalbehaviorassociates.com info@animalbehaviorassociates.com Aggressive behavior is one of the most common types of behavior problems seen by applied animal and veterinary behaviorists (Borchelt and Voith, 1985; Voith, 1981). While it may be frustrating to the owner, and even life-threatening to the animal when housesoiling, destructive and other annoying behavior problems occur, they are not associated with the potential risks, liabilities and dangerous situations as are aggressive behavior problems. Anywhere from 0.5 to 3.5 million people are bitten by dogs in the U.S. each year, and many dog bites go unreported. The most common victims are children. As many as 17 people are killed by dogs each year. Over 80% of the reported bites involve owned dogs, and most victims are familiar with the dogs that bite them (Wright, 1991). Although only 15% of reported bites involve dogs biting their owners, these are probably significantly under-reported. No statistics are available regarding the frequency of dog attacks on other dogs, although this is also a common type of aggressive problem. Dog owners often ask veterinarians about solutions to aggressive behavior problems, and veterinarians often encounter dogs displaying aggressive behavior in the clinic. Although some owners choose to surrender their dogs to shelters or have them euthanized because of aggression problems, others are so attached to their pets that having the dog leave the home may not be an option they are willing to consider, at least initially. Owners often describe their dogs as being "great dogs" except in particular, sometimes very narrowly defined contexts. In some cases, any injury the dog has inflicted is not severe, often not breaking the skin. What is Aggression? The terms "aggressive behavior" or "aggressive dog" tend to be used rather loosely. There is a great deal of confusion as to what is encompassed by the term "aggressive behavior". Dogs are said to be "aggressive" without identifying the criteria for placing them in that category. Since aggressive behavior is normal behavior for all dogs, should all dogs be labeled aggressive? If not, what does a dog have to do to be labeled as an "aggressive dog"? What is the difference between an "aggressive dog" and a "vicious dog"? Do behaviors such as growling, baring teeth, lunging, etc. fall into the category of aggressive behavior? Behaviorists identify a behavioral system known as agonistic behavior. Agonistic behaviors are those behaviors that are associated with competitive interactions or social conflict between animals. They include three large sub-categories of behavior:

10 1. aggressive behaviors which are intended to harm another individual such as biting and scratching. These can be either offensive or defensively motivated. 2. behaviors such as growling, snarling, baring teeth, air snapping, etc. (which can also be either offensive or defensive) that may precede aggressive behavior and are intended to threaten or warn another individual that an attack is likely, and 3. behaviors such as submission or escape which are intended to appease the aggressor and stop or prevent the attack It is often the case that dogs have displayed threatening behaviors without displaying aggressive behavior as defined above. Aggression: Understandable vs. Acceptable When aggressive behavior is elicited in response to environmental stimuli (as compared to CNS or other physiological abnormalities, for example), from the dog's perspective, the behavior is provoked. Provocation is usually evaluated based on the intent of the victim's behavior (e.g. attempt to be friendly vs. teasing or hurting the animal). However, the dog may perceive the victim's behavior differently than how it was intended. A common example is a child who hugs a dog as a display of affection. The dog may either be very frightened by this close contact and restraint or perceive the gesture as a dominance threat. From the dog's point of view, the aggression was provoked. But because people do not expect dogs to bite when being hugged, from the victim's point of view it was an unprovoked attack. By the same token, dogs aggressively guard their food when people approach too closely. Because owners are more likely to expect and understand this response, they may view the behavior as "instinctive" (a commonly misunderstood term) and therefore provoked and even acceptable. Because the concept of provocation is confounded by anthropomorphic interpretations of animal behavior, it has limited value in assessing aggression cases. Provocation is a separate issue from categorizing the behavior as acceptable. In both cases described above, the aggression is understandable, but it is not acceptable. "Vicious Dogs" Vicious is not a scientific term and will not be found in the ethological literature. Suggested behavioral criteria for vicious behavior include 1) aggressive behavior which is displayed in response to a wide variety of stimuli, 2) the aggressive behavior occurs with little or no warning threats, 3) the attack is at full intensity and not inhibited and 4) the behavior is not easily stopped. The legal definition of "vicious behavior" is very variable and often not congruent with this suggested behavioral one. For example, in Ohio a vicious dog is defined as: a dog that, without provocation, has killed or caused serious injury to any person; a dog that, without provocation, has caused injury, other than death or serious injury, to any person, or has killed another dog; or a pit bull.

11 Evaluating Dogs Displaying Aggressive or Threatening Behavior Very little data exist that measure the reliability and validity of any testing procedures that attempt to predict future aggressive behavior of a dog in the home environment. One study that has attempted to do so was conducted in 1991 in the Netherlands on dogs in an animal shelter. Results from this study found that the objective tests were better at predicting aggression problems than opinions of the shelter staff. Although sample sizes were small, reaction to a life-size doll successfully predicted aggression toward children in 5 of 7 cases. Possessive aggression over food or a bone and aggression toward adults in response to a threatening gesture were correctly predicted less than half of the time. Puppy tests that are supposed to predict adult behavior are not reliable. Because so little data exists on predicting aggression, a what you see is what you get approach may have to suffice. Dogs displaying threatening or aggressive behavior have made a choice to do so, despite other alternatives available to them (being submissive, leaving or avoiding the situation). If the environment is such that other choices are not possible, this may make an aggressive or threatening response more likely. Potential for Resolving Existing Aggression Problems Insufficient data exist to reliably predict the likelihood for successful resolution of aggression cases based on the behavioral history. Reisner et al. (1994) described several factors that were correlated with owners electing to euthanize dominantly aggressive dogs. However, this study did not evaluate the degree of behavior change or the efficacy of behavior modification. The following ideas for risk assessment are helpful, but not scientifically validated. Risk Assessment*: _ How dangerous is the behavior? Is the dog causing injury, only showing threatening behavior, air-snapping, or inhibited bites that result in minor or no injuries? _ How much overall control do the owners have of the dog? Is s/he responsive to basic training such as sit/come/down/stay? _ How generalized is the behavior? Is it limited to a few well-defined triggers or occurs in many contexts? _ How easily is the behavior triggered? Does the dog respond intensely to relatively minor triggers? e.g. will he lunge and growl just in response to eye contact or everyday handling, or is he aggressive or threatening only in response to intense threats such as physical punishment or a child falling on him? _ Is the dog s behavior unpredictable? Will he sometimes tolerant an event and at other times react with threats or aggression to the exact same set of circumstances? _ Is this a long-standing, chronic problem, or a recent, significant change in behavior? _ Is it difficult to get the dog to stop the threatening or aggressive behavior? _ Does the dog provide warning signals prior to snapping or biting? (owners may need to be educated about what to look for, as they may be unaware of the dog s warnings) _ Are any family members afraid of the dog? _ What about family factors? Are there children in the family? Is the family s lifestyle unpredictable and chaotic? How responsible and committed is the family to the dog? _ Is the dog an intact male? What is the size of the dog?

12 _ What is the dog s behavioral history? Is he in general well socialized? Is he mostly friendly or mostly unfriendly? How well does he tolerate being touched, handled, petted? * Modified from Pet Behavior Protocols: What To Say, What To Do, When to Refer. By S. Hetts, AAHA Press, Lakewood, CO, 1999. Literature Cited and Suggested Readings Borchelt, P.L. and V.L. Voith. 1985. Aggressive behavior in dogs and cats. Compend. Contin. Educ. Pract. Vet. 3:730-738. Borchelt, P.L. and V.L. Voith. 1986. Dominance aggression in dogs. Comp. Contin. Educ. Pract. Vet. 8: 36-44. Crowell-Davis, S. L. 1991. Identifying and correcting human-directed dominance aggression of dogs. Vet. Med. 86: 990-999. Domjan, M. and B. Burkhard. 1986. The Principles of Learning and Behavior, 2nd Edition. Brooks/Cole Publ. Company, Pacific Grove, CA. Marder, A. R. 1991. Psychotropic drugs and behavioral therapy. pp. 329-342 in Marder, A. R. and V. Voith, eds. Advances in companion animal behavior, Vet. Clinics of North America, Small Animal Practice 21 (2), W. B. Saunders, Phil. Overall, K.L. 1994. Prevention of aggressive disorders. Canine Practice 19: 19-22. Reisner, I. R., Erb, H. N., and K. A. Houpt. 1994. Risk factors for behavior-related euthanasia among dominant-aggressive dogs: 110 cases (1989-1992). JAVMA 205: 855-863. Van der Bor, J. et al. 1991. Behavioural testing of dogs in animal shelters to predict problem behaviour. Appl. Anim. Beh. Sci. 32: 237-251. Voith, V.L. 1981. Profile of 100 animal behavior cases. Mod. Vet. Pract. 483-484, June issue. Wright, J.C., 1991. Canine aggression toward people: bite scenarios and prevention. pp. 299-314 in Marder, A. R. and V. Voith, eds. Advances in companion animal behavior, Vet. Clinics of North America, Small Animal Practice 21 (2), W. B. Saunders, Phil. Young, M. S. 1982. Treatment of fear-induced aggression in dogs. pp. 645-654 in V.L. Voith and P.L. Borchelt, eds. Animal Behavior, The Vet. Clinics of North America, Small Animal Practice 12(4), W. B. Saunders, Phil.